Answered
Create an ROC Curve From Binary Classification Data
From |perfcurve| doc: *[X,Y] = perfcurve(labels,scores,posclass) computes a ROC curve for a vector of classifier predictions ...

13 years ago | 1

| accepted

Answered
Viewing the tree used for prediction by the method of "RUSBoost' in fitensemble
The strong learner is the ensemble. An ensemble is a collection of trees. It predicts by averaging predictions from individual t...

13 years ago | 1

| accepted

Answered
SVM classification with different kernels
See this thread <http://www.mathworks.com/matlabcentral/answers/57232-how-to-use-svmtrain-with-a-custom-kernel-in-matlab>

13 years ago | 0

Answered
Two linear discriminant functions to classify objects
I don't understand your question. But perhaps the following will help. In 2011b, we shipped |ClassificationDiscriminant| class w...

13 years ago | 0

Answered
How to use svmtrain() with a custom kernel in Matlab?
By convention adopted for |svmtrain|, observations are in rows and predictors are in columns. The same convention would hold for...

13 years ago | 0

Answered
[Compute AUC of a ROC Function]
If you have Statistics Toolbox, the |perfcurve| function can compute AUC. You mention that you have 3 classes. I am guessing ...

13 years ago | 0

Answered
Use of penalty multiplier C (boxconstraint) in svmtrain
Off hand I cannot think of a reason why you couldn't use |quadprog| to solve the 1-norm problem. I don't know why the 2-norm sol...

13 years ago | 0

Answered
Use of penalty multiplier C (boxconstraint) in svmtrain
The magnitude of the margin slack vector can be shrunk using a 1-norm or 2-norm penalty. The 1-norm version seems more popular, ...

13 years ago | 0

| accepted

Answered
Is there an LDA method that matches the format "D_t= W_Tc^T * E_t + w_0"
Sounds like you want a linear regression model: <http://www.mathworks.com/help/stats/linear-regression-model-workflow.html> ...

13 years ago | 0

Answered
hand gesture data classification and recognize gesture from image
Perhaps you could start by reading the User Guide for the Statistics Toolbox, where many tools for classification are collected:...

13 years ago | 0

Answered
model learning in the presence of noise
When you say "classifier ensemble", I assume you mean "decision tree ensemble". The time it takes to grow a tree depends on the ...

13 years ago | 0

Answered
How to run a chi-square test?
|chi2gof| tells you how closely the observed distribution resembles your theoretical guess. You don't have a distribution yet. T...

13 years ago | 0

Answered
Results of bootci and bootstrp funciton don't match.
You will get a better match to your quantile results if you run ci=bootci(1000, {@var, data}, 'type', 'per') Take a look...

13 years ago | 0

| accepted

Answered
Classregtree with multiple dependent variables
Your title says "multiple dependent variables", and your question says "a set of independent variables". Do you have multiple de...

13 years ago | 0

Answered
Algorithms for imbalanced multi class classification in Matlab?
I described approaches for learning on imbalanced data here <http://www.mathworks.com/matlabcentral/answers/11549-leraning-class...

13 years ago | 4

| accepted

Answered
How can I use LDA (Linear or Fisher Discrimnant Analysis) with an hardwritten digits dataset (like MNIST or USPS)?
I am not an expert in image analysis, but it seems you misunderstand what you need to do. LDA uses matrix X in which rows are ob...

13 years ago | 1

| accepted

Answered
How to report Terminal Node (only the one use to predict Y) Standard Deviation in TreeBagger?
TreeBagger is an ensemble of trees. All trees are different because they were grown on different bootstrap replicas of the data....

13 years ago | 0

| accepted

Answered
Sharing of a trained SVM model
There is no unified "SVM format" shared by different implementations. You can always dump the svmstruct to an ascii file and hop...

13 years ago | 0

| accepted

Answered
What is a good Dimensionality Reduction technique I can use?
Your best chance would be to set up variable selection based on that linear classifier you are using (you don't say what it is)....

13 years ago | 0

| accepted

Answered
Determining variables that contribute to principal components
The first paragraph in the doc description for princomp says "COEFF is a p-by-p matrix, each column containing coefficients for ...

13 years ago | 0

| accepted

Answered
How to find the important features from many features of a MFFC speech sample??
I described feature ranking/selection tools available from Statistics Tlbx here: <http://www.mathworks.com/matlabcentral/answers...

13 years ago | 0

Answered
Select machine learning features in Matlab
Here are Statistics Toolbox utilities you should look into: # sequentialfs # relieff # predictorImportance method of Classi...

13 years ago | 1

| accepted

Answered
fitensemble adaboost m1 stump picking
A decision tree in a boosting ensemble by default minimizes the Gini diversity index, not classification error. Two child nodes ...

13 years ago | 0

Answered
How to best determine the probability of a distribution given an outlying observation?
If you know the reference distribution analytically, you can compute its cdf at the smallest observed value. Suppose this cdf va...

13 years ago | 0

| accepted

Answered
fitensemble with adaboost and trees
http://www.mathworks.com/help/toolbox/stats/classificationensembleclass.html http://www.mathworks.com/help/toolbox/stats/comp...

13 years ago | 0

| accepted

Answered
Question about KNN and how to use it.
Take a look at the User Guide for k-NN classification: http://www.mathworks.com/help/toolbox/stats/bsehyju-1.html In your cas...

13 years ago | 0

| accepted

Answered
Why covariance matrix is computed like this?
m-1 in the denominator gives the unbiased estimate, and m gives the max likelihood estimate. http://en.wikipedia.org/wiki/Est...

13 years ago | 0

Answered
Matlab TreeBagger Cost argument not working as it works with similar function fitensemble
fitensemble and TreeBagger use different approaches to incorporate the cost matrix. fitensemble uses the cost matrix to updat...

13 years ago | 0

Answered
Help using perfcurve function!!
Other than a few typos (for example, y1 = (1:132)'>70; clearly is not what you meant), I don't see anything wrong with your code...

13 years ago | 0

| accepted

Answered
Help using perfcurve function!!
This is a rather vague question. You might get more replies if you explain what approaches you are thinking about and what you h...

13 years ago | 0